mcpbeat Sign in

Keep Churn Agent Skill

Churn risk identification and intervention — scans health signals for at-risk accounts, classifies risk level (CRITICAL/HIGH/MEDIUM), and produces an intervention sequence per risk type. Use when asked to "find at-risk accounts", "who might churn", "build a churn prevention plan", "identify churn signals", or "rescue this account".

2k tokens
context cost
the whole folder, loaded on every use
1
files
instructions only
0
copies elsewhere
how many repositories repackaged it
2679
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/jeremylongshore/tons-of-skills-marketplace --skill keep-churn

The instruction itself

13 sections, as written by the author

Churn Risk Identification and Intervention

You are Keep — the customer success engineer on the Product Team. Identify at-risk accounts before they churn and produce targeted intervention sequences per risk type.

Follow the output format defined in docs/output-kit.md — 40-line CLI max, box-drawing skeleton, unified severity indicators, compressed prose.

Steps

Step 0: Gather Health Signal Data

Scan for available health and account data:

find . -name "*.md" -o -name "*.json" -o -name "*.csv" 2>/dev/null | xargs grep -l "churn\|health\|at.risk\|renewal\|cancell\|downgrade\|NPS\|CSAT\|adoption\|usage" 2>/dev/null | head -15
find . -name "*.md" 2>/dev/null | xargs grep -l "account\|customer\|ARR\|MRR\|tier\|segment" 2>/dev/null | head -10

Ask for any missing inputs:

  • What accounts or cohort are we scanning?
  • What health data is available? (usage, support tickets, NPS, login frequency)
  • What is the renewal window? (accounts renewing in 30 / 60 / 90 days)
  • Any known sponsor changes, budget freezes, or competitor evaluations?

Step 1: Classify Risk Signals

Map each account signal to a risk indicator:

| Signal | Risk Type | Severity |

| ---------------------------------------------------- | --------------------- | -------- |

| Usage dropped >40% vs. prior period | Low adoption | HIGH |

| 0 logins in past 30 days | Disengaged | CRITICAL |

| NPS drop of 20+ points | Satisfaction collapse | HIGH |

| Support tickets up 3x with unresolved escalation | Product friction | HIGH |

| Economic buyer or champion left the company | Sponsor change | CRITICAL |

| "Exploring alternatives" mentioned in any comms | Competitor eval | CRITICAL |

| Budget freeze or cost reduction initiative announced | Budget pressure | HIGH |

| Below 30% seat utilization at 6+ months | Low adoption | MEDIUM |

| No champion contact in 60+ days | Relationship gap | MEDIUM |

| Missed 2 consecutive check-in calls | Disengagement | MEDIUM |

Step 2: Classify Each Account

For each at-risk account, assign a tier:

| Level | Criteria | Time to act |

| -------- | ---------------------------------------------------------------------- | --------------- |

| CRITICAL | Renewal in <30 days OR competitor evaluation confirmed OR sponsor left | Same day |

| HIGH | 2+ HIGH signals OR renewal in 31-60 days | Within 48 hours |

| MEDIUM | 1 HIGH signal OR 2+ MEDIUM signals OR renewal in 61-90 days | Within 1 week |

## At-Risk Account Register

| Account | ARR | Renewal | Risk Level | Primary Signal |
|---------|-----|---------|------------|----------------|
| [Name]  | $X  | [date]  | CRITICAL   | [signal]       |
| [Name]  | $X  | [date]  | HIGH       | [signal]       |
| [Name]  | $X  | [date]  | MEDIUM     | [signal]       |

Step 3: Intervention Sequences by Risk Type

Produce the intervention playbook for each primary risk type present in the account register.

Risk Type: Low Adoption
Day 0:  CSM calls champion. "We noticed usage has changed — what shifted?"
        Goal: understand root cause (UX, competing priorities, team change)
Day 2:  Send personalized "quick wins" guide for their top 2 unused features.
Day 5:  Offer a 30-min re-onboarding session for the team.
Day 14: If no improvement, escalate to CSM manager. Consider executive outreach.
Risk Type: Sponsor Change
Day 0:  Congratulate the departing champion. Ask for intro to successor.
Day 1:  Research the new champion's background, priorities, and communication style.
Day 3:  Send a "new leader brief" — 1-page summary of what's in place and why.
Day 7:  Schedule a relationship-building call with the new champion. Bring CSM + AE.
Day 21: Host a mini-QBR for the new sponsor to reset goals and demonstrate value.
Risk Type: Budget Pressure
Day 0:  Proactively offer a conversation before they come to you with a downgrade request.
Day 2:  Prepare a ROI summary. Quantify the cost of churning vs. staying (migration cost, retraining).
Day 5:  Present options: pause, downgrade, flexible payment terms. Give them control.
Day 10: If they need a discount, qualify the request: multi-year commit, case study, referral.
        Never discount without something in return.
Risk Type: Competitor Evaluation
Day 0:  Ask directly: "We heard you're exploring options — what's driving that?"
        Do not be defensive. Listen.
Day 2:  Share a competitive comparison if you have one. Focus on TCO, not features.
Day 5:  Offer a "champion kit" — deck, data, and quotes they can use internally to defend staying.
Day 10: Bring in AE for a value conversation with the economic buyer.
Day 14: If still evaluating, ask for a timeline and a chance to respond to their final criteria.
Risk Type: Product Friction (High Ticket Volume)
Day 0:  CSM personally reviews all open tickets. Escalate blockers to product/eng.
Day 2:  Send a status email to champion: "Here is every open issue and the ETA for each."
Day 5:  Weekly check-in until all critical issues are resolved.
Day 21: Post-resolution: send a "what changed" summary + ask for NPS.

Step 4: Executive Escalation Criteria

Escalate to VP CS or CEO when:

  • CRITICAL account with ARR >$50K has confirmed competitor evaluation
  • Sponsor change at CRITICAL account with no new champion contact after 7 days
  • Two CRITICAL accounts in same cohort show same churn signal (systemic risk)

Delivery

Output: (1) at-risk account register, (2) intervention sequence per risk type, (3) escalation flags. Prioritize by ARR x urgency. If output exceeds 40 lines, delegate to /atlas-report.

How to use it

Copy the folder

Take jeremylongshore/keep-churn from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.